The observations for the validation were taken from 2015-01-29 to 2023-12-03.
Remember that we have on a daily scale the variables
sfcWind, tas, pr,
tasmax, tasmin and psl and in a
monthly scale clt, rsdt,rsds as a
predictors. We also have the month and the daily daylight amount in
seconds as a predictors too.
| Metric | xgboost | cnn | naive | lstm |
|---|---|---|---|---|
| mae | 12.07 | 14.42 | 14.54 | 12.34 |
| cor | 0.84 | 0.77 | 0.76 | 0.82 |
| ratio_of_sd | 0.81 | 0.81 | 0.76 | 0.86 |
| KGE | 0.75 | 0.68 | 0.67 | 0.78 |
| amplitude_mae_monthly | 16.11 | 16.83 | 16.97 | 13.45 |
| sign_correlation | 0.75 | 0.72 | 0.70 | 0.75 |
| extreme_correlation | 0.53 | 0.44 | 0.52 | 0.58 |
| qqplot_mae | 4.80 | 5.40 | 7.48 | 3.78 |
| acf_mae | 0.02 | 0.04 | 0.05 | 0.02 |
| extremogram_mae | 0.02 | 0.03 | 0.02 | 0.02 |
Important: Right now we are only estimating the upper tail
extremogram. Currently we didn’t find a way to estimate the two tales at
the same time. We are using quant = .97